View Source Bumblebee.Vision.DinoV2 (Bumblebee v0.5.3)
DINOv2 model family.
Architectures
:base
- plain DINOv2 without any head on top:for_image_classification
- DINOv2 with head for image classification:backbone
- DINOv2 with feature maps output
Inputs
"pixel_values"
-{batch_size, image_size, image_size, num_channels}
Featurized image pixel values.
"patch_mask"
-{batch_size, num_patches}
Mask to nullify selected embedded patches.
Configuration
:image_size
- the size of the input spatial dimensions. The model is trained for this size, however the model supports any other input size by interpolating position embeddings . Defaults to518
:num_channels
- the number of channels in the input. Defaults to3
:patch_size
- the size of the patch spatial dimensions. Defaults to14
:hidden_size
- the dimensionality of hidden layers. Defaults to384
:num_blocks
- the number of Transformer blocks in the encoder. Defaults to12
:num_attention_heads
- the number of attention heads for each attention layer in the encoder. Defaults to12
:intermediate_size_ratio
- the dimensionality of the intermediate layer in the transformer feed-forward network (FFN) in the encoder, expressed as a multiplier of:hidden_size
. Defaults to4
:use_qkv_bias
- whether to use bias in query, key, and value projections. Defaults totrue
:activation
- the activation function. Defaults to:gelu
:ffn_swiglu_activation
- whether to use the gated SwiGLU activation function in the feed-forward network (FFN). Defaults tofalse
:scale_initial_value
- the initial value for scaling layers. Defaults to1.0
:dropout_rate
- the dropout rate for encoder and decoder. Defaults to0.0
:attention_dropout_rate
- the dropout rate for attention weights. Defaults to0.0
:layer_norm_epsilon
- the epsilon used by the layer normalization layers. Defaults to1.0e-6
:initializer_scale
- the standard deviation of the normal initializer used for initializing kernel parameters. Defaults to0.02
:backbone_output_indices
- list of indices indicating which feature maps to include in the output. If not specified, only the last feature map is included:backbone_use_norm
- whether to add layer normalization layer to each of the feature maps returned by the backbone. Defaults totrue
:output_hidden_states
- whether the model should return all hidden states. Defaults tofalse
:output_attentions
- whether the model should return all attentions. Defaults tofalse
:num_labels
- the number of labels to use in the last layer for the classification task. Defaults to2
:id_to_label
- a map from class index to label. Defaults to%{}